| Message ID | 20240617105311.1587489-3-a-bhatia1@ti.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | drm/bridge: cdns-dsi: Fix the color-shift issue | expand |
On Mon, Jun 17, 2024 at 04:23:03PM GMT, Aradhya Bhatia wrote: > Update the Phy initialized state to "not initialized" when the driver > (and the hardware by extension) gets suspended. This will allow the Phy > to get initialized again after resume. > > Fixes: e19233955d9e ("drm/bridge: Add Cadence DSI driver") > Signed-off-by: Aradhya Bhatia <a-bhatia1@ti.com> > --- > drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c > index b016f2ba06bb..42565e253b2d 100644 > --- a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c > +++ b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c > @@ -1153,6 +1153,7 @@ static int __maybe_unused cdns_dsi_suspend(struct device *dev) > clk_disable_unprepare(dsi->dsi_p_clk); > reset_control_assert(dsi->dsi_p_rst); > dsi->link_initialized = false; Most likely you should also call phy_exit() here. And in _remove() too. > + dsi->phy_initialized = false; > return 0; > } > > -- > 2.34.1 >
Hi Dmitry, Thanks for reviewing the patches! On 17/06/24 17:29, Dmitry Baryshkov wrote: > On Mon, Jun 17, 2024 at 04:23:03PM GMT, Aradhya Bhatia wrote: >> Update the Phy initialized state to "not initialized" when the driver >> (and the hardware by extension) gets suspended. This will allow the Phy >> to get initialized again after resume. >> >> Fixes: e19233955d9e ("drm/bridge: Add Cadence DSI driver") >> Signed-off-by: Aradhya Bhatia <a-bhatia1@ti.com> >> --- >> drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c | 1 + >> 1 file changed, 1 insertion(+) >> >> diff --git a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c >> index b016f2ba06bb..42565e253b2d 100644 >> --- a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c >> +++ b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c >> @@ -1153,6 +1153,7 @@ static int __maybe_unused cdns_dsi_suspend(struct device *dev) >> clk_disable_unprepare(dsi->dsi_p_clk); >> reset_control_assert(dsi->dsi_p_rst); >> dsi->link_initialized = false; > > Most likely you should also call phy_exit() here. And in _remove() too. I agree that phy_exit should be called here. But why in _remove()? Wouldn't having phy_exit in 2 places mess up the internal ref count? > >> + dsi->phy_initialized = false; >> return 0; >> } >> >> -- >> 2.34.1 >> > -- Regards Aradhya
On Mon, 17 Jun 2024 at 17:16, Aradhya Bhatia <a-bhatia1@ti.com> wrote: > > Hi Dmitry, > > Thanks for reviewing the patches! > > On 17/06/24 17:29, Dmitry Baryshkov wrote: > > On Mon, Jun 17, 2024 at 04:23:03PM GMT, Aradhya Bhatia wrote: > >> Update the Phy initialized state to "not initialized" when the driver > >> (and the hardware by extension) gets suspended. This will allow the Phy > >> to get initialized again after resume. > >> > >> Fixes: e19233955d9e ("drm/bridge: Add Cadence DSI driver") > >> Signed-off-by: Aradhya Bhatia <a-bhatia1@ti.com> > >> --- > >> drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c | 1 + > >> 1 file changed, 1 insertion(+) > >> > >> diff --git a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c > >> index b016f2ba06bb..42565e253b2d 100644 > >> --- a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c > >> +++ b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c > >> @@ -1153,6 +1153,7 @@ static int __maybe_unused cdns_dsi_suspend(struct device *dev) > >> clk_disable_unprepare(dsi->dsi_p_clk); > >> reset_control_assert(dsi->dsi_p_rst); > >> dsi->link_initialized = false; > > > > Most likely you should also call phy_exit() here. And in _remove() too. > > I agree that phy_exit should be called here. But why in _remove()? > Wouldn't having phy_exit in 2 places mess up the internal ref count? If suspend() is going to be called in the teardown path, then it's fine to have just one call here. Otherwise you might add one guarded with if (phy_initialized) to _remove() too. > > > > >> + dsi->phy_initialized = false; > >> return 0; > >> } > >> > >> -- > >> 2.34.1 > >> > > > -- > Regards > Aradhya
On 18/06/24 01:04, Dmitry Baryshkov wrote: > On Mon, 17 Jun 2024 at 17:16, Aradhya Bhatia <a-bhatia1@ti.com> wrote: >> >> Hi Dmitry, >> >> Thanks for reviewing the patches! >> >> On 17/06/24 17:29, Dmitry Baryshkov wrote: >>> On Mon, Jun 17, 2024 at 04:23:03PM GMT, Aradhya Bhatia wrote: >>>> Update the Phy initialized state to "not initialized" when the driver >>>> (and the hardware by extension) gets suspended. This will allow the Phy >>>> to get initialized again after resume. >>>> >>>> Fixes: e19233955d9e ("drm/bridge: Add Cadence DSI driver") >>>> Signed-off-by: Aradhya Bhatia <a-bhatia1@ti.com> >>>> --- >>>> drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c | 1 + >>>> 1 file changed, 1 insertion(+) >>>> >>>> diff --git a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c >>>> index b016f2ba06bb..42565e253b2d 100644 >>>> --- a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c >>>> +++ b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c >>>> @@ -1153,6 +1153,7 @@ static int __maybe_unused cdns_dsi_suspend(struct device *dev) >>>> clk_disable_unprepare(dsi->dsi_p_clk); >>>> reset_control_assert(dsi->dsi_p_rst); >>>> dsi->link_initialized = false; >>> >>> Most likely you should also call phy_exit() here. And in _remove() too. >> >> I agree that phy_exit should be called here. But why in _remove()? >> Wouldn't having phy_exit in 2 places mess up the internal ref count? > > If suspend() is going to be called in the teardown path, then it's > fine to have just one call here. Otherwise you might add one guarded > with if (phy_initialized) to _remove() too. Since the _suspend() and _resume() hooks are registered for both, system sleep PM and runtime PM, the _suspend() hook does get called when I remove the cdns_dsi module using modprobe. Tested that on my setup. I believe having phy_exit() only in _suspend() would suffice for now. > >> >>> >>>> + dsi->phy_initialized = false; >>>> return 0; >>>> } >>>> >>>> -- >>>> 2.34.1 >>>> >>> >> -- >> Regards >> Aradhya > > > -- Regards Aradhya
diff --git a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c index b016f2ba06bb..42565e253b2d 100644 --- a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c +++ b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c @@ -1153,6 +1153,7 @@ static int __maybe_unused cdns_dsi_suspend(struct device *dev) clk_disable_unprepare(dsi->dsi_p_clk); reset_control_assert(dsi->dsi_p_rst); dsi->link_initialized = false; + dsi->phy_initialized = false; return 0; }
Update the Phy initialized state to "not initialized" when the driver (and the hardware by extension) gets suspended. This will allow the Phy to get initialized again after resume. Fixes: e19233955d9e ("drm/bridge: Add Cadence DSI driver") Signed-off-by: Aradhya Bhatia <a-bhatia1@ti.com> --- drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c | 1 + 1 file changed, 1 insertion(+)